Maxim Cookbooks

Maxim Cookbooks

Maxim Cookbooks is a collection of example projects demonstrating how to integrate [**Maxim**](https://www.getmaxim.ai/) for AI agent **observability**, **auto-evaluation**, **prompt management**, **simulation**, and **test run workflows** across a variety of popular frameworks and SDKs.

Maxim is an end-to-end AI evaluation and observability platform that empowers modern AI teams to ship agents with quality, reliability, and speed.

🔧 What’s in this Repository?

This repository provides ready-to-run **code samples**, **notebooks**, and **configuration files** to help you quickly understand how to:

• Integrate observability and tracing in agents.
• Perform automated evaluations on agent outputs.
• Manage and retrieve prompts via Maxim APIs.
• Simulate agent behavior with pre-defined workflows.
• Run offline and hosted test runs for evaluation and regression tracking.

👇Directory Structure

maxim-cookbooks/
├── .github/                 # GitHub action workflows
├── .vscode/                # Editor configuration
├── go/
│   └── observability-online-eval/
│       ├── azure/
│       ├── bedrock/
│       └── openai/
├── java/                   # (WIP or initial structure)
├── python/
│   ├── observability-online-eval/
│   ├── prompt-management/
│   ├── simulation/
│   └── test-runs/
├── typescript/
│   ├── langchain/
│   ├── langgraph/customer-support/
│   ├── llamaindex/
│   └── test-runs/
├── README.md

🧩 Some Use Case Examples

1. Observability in Agents

Crew AI Agent with Maxim

Add observability to a [Crew AI agent](https://github.com/maximhq/maxim-cookbooks/blob/main/python/observability-online-eval/crew-ai/cooking-agent/agent.py):

from maxim import Config, Maxim
from maxim import logger
from maxim.logger import LoggerConfig
from maxim.maxim import LoggerConfig
from maxim.logger.langchain import MaximLangchainTracer

maxim_api_key = os.environ.get("MAXIM_API_KEY", "")
maxim_base_url